31.5m art images online?
April 2 2015
Picture: TAN
Interesting story in The Art Newspaper by Martin Bailey, who reports that an 'International Digital Photo Archive Consortium' is mulling over plans to digitise their entire collections of archival images. That is, images of paintings in exhibition catalogues and sale catalogues since - essentially - photography began. The Consortium includes the Witt library in London, and the Frick in New York, among others. The total number of combined images is apparently some 31.5m.
Martin writes:
Many of these archives still mount images of works with captions on thin card, filed by artist, in alphabetical order. Each artist work is subdivided by type—for example portraits, landscapes and still lifes. Most have not been digitised, so researchers have to visit the library in person.
The plan is to digitise the 31.5 million cards held by 14 of the world’s leading archives and then upload them on the web to make them easily searchable. No decisions have been made on what would be available free or for a charge. The images would be for research purposes, rather than reproduction.
Chris Stolwijk, director of the Netherlands Institute for Art History, says that it is “essential to go digital, otherwise we will be working for a only very small group of researchers”. His institute’s collection is the largest in the world, with 7 million images. Amassed since the 1930s, it is particularly strong in Netherlandish and Dutch Golden Age paintings.
The RKD alwready leads the way in this area, with a large number of digital images. The Frick has a good image library, but it's approach to online - at the moment - is maddening: it has a good database of what images it does have, but it's unillustrated, and if you ask them for an image, they will only send you one in the post! Perhaps the problem here is copyright, which I suspect will be a pretty difficult barrier to overcome. Recent regulations in the UK, for example, have made such projects all but impossible. And although the Witt Library just about still exists, funding cuts implemented by the Courtauld Institute means that it has stopped collecting images, and has no specialist staff.
But if it could be made to work, the benefits on an online database like this would be extraordinary. And if the images were married up to, say, something like Google image search, then anyone wanting to know what an unidentifed painting was could easily find out what it was, who it was by, or where it had been, just by running a search - as long as the painting had been photographed before at some point in its life.
And since most paintings have been photographed at some point, in an earlier sale for example, then those who rely on their visual memory to make a living by actually knowing such things, like me, would be out of a job. Picture sleuthing would become a thing of the past. I'm not going to say this is a Bad Thing at all - progress is inevitable, and it'll be just one more thing that the computers can do better than us. But until then, I suppose I must make hay while I can...
